‘What answer then is to be given to them?’ said Apollonides, ‘for the difficulty about reflexion seems to apply equally to us.’ ‘Equally no doubt in one sense,’ I answered, ‘but in another sense not at all so. First look at the details of the simile, how “topsy turvy”[351] it is, rivers flowing up stream! The water is below and on earth, the moon is above the earth and poised aloft. So the angles of reflexion are differently formed; in the one case the apex is above in the moon, in the other below on the earth. They should not then require that mirrors should produce every image and like reflexions at any distance, since [Sidenote: E] they are fighting against clear fact. But from those like ourselves who seek to show that the moon is not a fine smooth substance like water, but heavy and earth-like, it is strange to ask for a visible appearance of the sun in her. Why, milk does not return such mirrored images, nor produce optical reflexion, the reason being the unevenness and roughness of its parts. How can the moon possibly send back the vision off [Sidenote: F] herself as the smoother mirrors do? We know that even in these, if any scratch or speck or roughness is found at the point from which the vision is naturally reflected it is obscured; the blemishes are seen, but they do not return the light. A man who requires that she should either turn our vision back to the sun, or else not reflect the sun from herself to us, is a humorist; he wants our eye to be the sun, the image light, man heaven! That the reflexion of the sun’s light conveyed to the moon with the impact of his intense brilliance should be borne back to us is reasonable enough, whereas our sight is weak and slight and merely fractional. What wonder if it deliver a stroke which has no resilience, or, if it does rebound, no continuity, but is broken up and fails, having no store of light to make up for dispersion about the rough and uneven [Sidenote: 937] places. For it is not impossible that the reflexion should rebound to the sun from water and other mirrors, being still strong and near its point of origin; whereas from the moon, even if there are glancings of a sort, yet they will be weak and dim, and will fail by the way because of the long distance. Another point: concave mirrors return the reflected light in greater strength than the original, and thus often produce [Sidenote: B] flames; convex and spherical mirrors one which is weak and dim, because the pressure is not returned from all parts of the surface. You have seen, no doubt, how when two rainbows appear, one cloud enfolding another, the enveloping bow shows the colours dim and indistinct, for the outer cloud lying further from the eye does not return the reflexion in strength or intensity. But enough! Whereas the light of the sun reflected from the moon loses its heat entirely, and only a scanty and ineffectual remnant of its brilliance reaches us, do you really think it